So anyone with deep enough pockets can do it.

However, just because you receive a fine does not mean that you "can't" do it. You've already done it, got caught, now a fine. It does not mean that the LLM model has to be tossed out and destroyed with a new version trained up without that data. It just means can't is a very stupid word to imply here.

Were these from the same high-profile publishers?

What was the judgment? Seems that their domains are still active. Why is there a difference in judgment here?